The long-term objective is to understand the role of recombination processes and DNA mismatch repair in the ability of cells to survive drug and radiation-induced DNA damage. We have obtained data with various agents with different mechanisms of action suggesting that DNA double-strand breaks are a common denominator in cells exposed to DNA damaging agents. These agents include not only those occurring naturally (nitric oxide), those used in the therapy of neoplastic disease (e.g., cisplatin, streptozotocin) but also genotoxicants in our environment (e.g.,methylating agents). Ionizing radiation is known to produce double-strand breaks directly but the other agents do not. We hypothesize that drug-induced double-strand breaks are formed by replication through DNA containing single-strand nicks produced by repair processes leading to collapse of the replication fork. Recombination mechanisms are required to repair these double-strand breaks and are therefore extremely important for cellular defenses against these agents in combination with other repair pathways and replication restart proteins. To test our hypothesis we propose to use E. coli as a model system as more is known about DNA replication, repair and recombination than any other organism. Furthermore, mutant strains affecting these processes are available only in this organism. DNA mismatch repair is known to sensitize human and E. coli cells to the toxic effects of cisplatin and methylating agents. Indeed tumor cells resistant to these agents have been isolated from patients and shown to be deficient in mismatch repair. We hypothesize that the role of mismatch repair in drug- resistance is because it no longer interferes with recombinational repair of drug-induced damage. The first specific aim will test the hypothesis that double-strand breaks are formed in chromosomal DNA after exposure to nitric oxide, methylating agents and cisplatin using gamma-radiation as the control. We will use physical methods to detect such breaks. In the second aim, a simple recombination assay will be used to monitor the level of double-strand breaks induced by the agents listed above and by gamma radiation. We will also determine the gene products required for drug-induced recombination. The third aim will use a biochemical assay with purified proteins and substrates to test the hypothesis that mismatch repair interferes with recombination. If these experiments support our model, it will open up a new approach to combat tumor cells by targeting recombination proteins.